An image registration method, apparatus, device and medium

Abstract

This invention discloses an image registration method, apparatus, device, and medium. The method includes: performing a first-stage registration between a non-reference image in a group of images to be registered and a reference image in the group, resulting in a preliminary registered image group; segmenting the images in the group of images to be registered to obtain a segmentation result of the target anatomical structure in the calibration space of the reference image; overlaying the segmentation result with each image in the preliminary registered image group and obtaining the region of interest; performing region registration between the region of interest of the non-reference images in the preliminary registered image group and the region of interest of the reference image; and performing a second-stage registration between the non-reference images in the preliminary registered image group based on the registration relationship during the region registration process. The technical solution of this invention improves the registration effect of local image regions, thereby increasing the accuracy of region of interest analysis.

Description

[0001] This application is a divisional application of application number 2021105928825. Technical Field

[0002] The embodiments of the present invention relate to the field of medical image processing technology, and in particular to an image registration method, apparatus, device and medium. Background Technology

[0003] Existing image registration methods automatically execute pixel-based registration algorithms on the image pairs to be registered, obtaining the registration relationship between two or more images. Alternatively, users can manually register the images by performing manual operations such as translation, rotation, and scaling on the images in the image pair to achieve spatial alignment. Alternatively, the image registration device can automatically calculate the registration relationship between two point sets or two ROI images based on information input by the user on the interface, such as selecting paired point sets or selecting regions of interest (ROIs), and finally display the registration alignment status.

[0004] However, the above registration methods are all for image information based on the global or regional field of view of the image data. Due to the large distribution range and tortuous direction of blood vessels, the different imaging times of multiple sequences, the changes in the microenvironment of blood vessels, and the limb movement of the scanned object, the registration relationship of blood vessels between multiple sequences is complex and does not satisfy the global rigid transformation relationship. Therefore, the above registration algorithms cannot achieve good registration results. Summary of the Invention

[0005] This invention provides an image registration method, apparatus, device, and medium to improve the registration effect of local areas of an image, thereby increasing the accuracy of region of interest analysis.

[0006] In a first aspect, embodiments of the present invention provide an image registration method, the method comprising:

[0007] A group of images to be registered is obtained, and the non-reference images in the group of images to be registered are registered to the reference images in the group of images to be registered in the first stage to obtain a preliminary registered image group.

[0008] Perform target anatomical structure segmentation on any image in the group of images to be registered to obtain the segmentation result of the target anatomical structure in the calibration space where the reference image is located;

[0009] The segmentation results are overlaid with each image in the preliminary registered image group to obtain the region of interest in the target anatomical structure.

[0010] The regions of interest (ROIs) of the non-reference images in the preliminary registered image group are registered with the ROI of the reference image. Then, based on the registration relationship in the ROI process, the non-reference images in the preliminary registered image group are registered in a second stage to complete the image registration process.

[0011] Optionally, the first-stage registration of the non-reference images in the image group to be registered to the reference images in the image group includes:

[0012] The registration relationship between the non-reference image and the reference image in the image group to be registered is determined according to a preset image registration algorithm;

[0013] The non-reference images in the image group to be registered are reconstructed according to the registration relationship.

[0014] Optionally, when the object of image segmentation is a non-reference image in the image group to be registered, the step of segmenting the target anatomical structure in any image in the image group to be registered to obtain the segmentation result of the target anatomical structure in the calibration space where the reference image is located includes:

[0015] A preset image segmentation algorithm is used to segment the target anatomical structure of the non-reference image in the image group to be registered, and the segmentation result of the target anatomical structure is extracted.

[0016] Based on the registration relationship in the first stage registration process, the segmentation result is reconstructed to obtain the segmentation result of the target anatomical structure in the calibration space where the reference image is located.

[0017] Optionally, obtaining the region of interest in the target anatomical structure includes:

[0018] A preset target region recognition algorithm is used to detect the region of interest in the target anatomical structure.

[0019] Optionally, obtaining the region of interest in the target anatomical structure further includes:

[0020] In response to an image selection operation in the image display area, a region of interest is determined in the target anatomical structure.

[0021] Optionally, the preset image registration algorithm includes:

[0022] Methods based on pixel coordinate information calculation, and medical image registration algorithms based on pixel grayscale information or image features.

[0023] Optionally, region registration is performed between the regions of interest (ROIs) of the non-reference images in the preliminary registered image group and the ROI of the reference image, and a second-stage registration is performed on the non-reference images in the preliminary registered image group according to the registration relationship in the region registration process, thus completing the image registration process, including:

[0024] By using a preset medical image registration algorithm, the region registration relationship between the region of interest of the non-reference image in the preliminary registration image group and the region of interest of the reference image is determined. Then, based on the region registration relationship, the non-reference image in the preliminary registration image group is registered in the second stage to complete the image registration process.

[0025] Secondly, embodiments of the present invention also provide an image registration device, the device comprising:

[0026] The first registration module is used to acquire a group of images to be registered, and to perform a first-stage registration of the non-reference images in the group of images to be registered to the reference images in the group of images to be registered, so as to obtain a preliminary registered image group.

[0027] The structural segmentation module is used to segment the target anatomical structure of any image in the group of images to be registered, and to obtain the segmentation result of the target anatomical structure in the calibration space where the reference image is located.

[0028] The target region selection module is used to overlay the segmentation results with each image of the preliminary registered image group and obtain the region of interest in the target anatomical structure.

[0029] The second registration module is used to perform region registration between the region of interest of the non-reference image in the preliminary registration image group and the region of interest of the reference image, and to perform a second-stage registration of the non-reference image in the preliminary registration image group according to the registration relationship in the region registration process, thereby completing the image registration process.

[0030] Optionally, the first registration module is specifically used for:

[0031] The registration relationship between the non-reference image and the reference image in the image group to be registered is determined according to a preset image registration algorithm;

[0032] The non-reference images in the image group to be registered are reconstructed according to the registration relationship.

[0033] Optionally, when the object of image segmentation is a non-reference image in the group of images to be registered, the structure segmentation module is specifically used for:

[0034] A preset image segmentation algorithm is used to segment the target anatomical structure of the non-reference image in the image group to be registered, and the segmentation result of the target anatomical structure is extracted.

[0035] Based on the registration relationship in the first stage registration process, the segmentation result is reconstructed to obtain the segmentation result of the target anatomical structure in the calibration space where the reference image is located.

[0036] Optionally, the target area selection module is used for:

[0037] A preset target region recognition algorithm is used to detect the region of interest in the target anatomical structure.

[0038] Optionally, the target area selection module is further configured to:

[0039] In response to an image selection operation in the image display area, a region of interest is determined in the target anatomical structure.

[0040] Optionally, the preset image registration algorithm includes:

[0041] Methods based on pixel coordinate information calculation, and medical image registration algorithms based on pixel grayscale information or image features.

[0042] Optionally, the second registration module is specifically used for:

[0043] By using a preset medical image registration algorithm, the region registration relationship between the region of interest of the non-reference image in the preliminary registration image group and the region of interest of the reference image is determined. Then, based on the region registration relationship, the non-reference image in the preliminary registration image group is registered in the second stage to complete the image registration process.

[0044] Thirdly, embodiments of the present invention also provide a computer device, the computer device comprising:

[0045] One or more processors;

[0046] Storage device for storing one or more programs;

[0047] When the one or more programs are executed by the one or more processors, the one or more processors implement any of the image registration methods described in the embodiments of the present invention.

[0048] Fourthly, embodiments of the present invention also provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the image registration method as described in any of the embodiments of the invention.

[0049] In this embodiment of the invention, a preliminary registered image group is obtained by performing a first-stage registration of non-reference images in a group of images to be registered to a reference image, achieving spatial alignment at the macroscopic structural level. Then, the target anatomical structure is segmented in any image of the group of images to be registered, yielding a segmentation result of the target anatomical structure in the calibration space of the reference image. The segmentation result is then overlaid with each image in the preliminary registered image group to obtain the region of interest (ROI) within the target anatomical structure. Region registration is performed between the ROI of the non-reference images in the preliminary registered image group and the ROI of the reference image. A second-stage registration is then performed on the non-reference images in the preliminary registered image group based on the registration relationship established during the region registration process, completing the image registration process. This embodiment of the invention solves the problem of poor registration results in local regions that do not meet rigidity requirements; it can improve the registration effect of local image regions, thereby increasing the accuracy of ROI analysis. Attached Figure Description

[0050] Figure 1 This is a flowchart of the image registration method in Embodiment 1 of the present invention;

[0051] Figure 2 This is a schematic diagram of the image group to be registered in Embodiment 1 of the present invention;

[0052] Figure 3 This is a schematic diagram of the preliminary registered image group after the first stage of registration in Embodiment 1 of the present invention;

[0053] Figure 4 This is a schematic diagram of the region of interest selection in Embodiment 1 of the present invention;

[0054] Figure 5 These are comparison images of the registration results of the region of interest in Embodiment 1 of the present invention after two stages;

[0055] Figure 6 These are comparison images of the registration results of the region of interest in Embodiment 1 of the present invention after two stages;

[0056] Figure 7 This is a schematic diagram of the image registration device in Embodiment 2 of the present invention;

[0057] Figure 8 This is a schematic diagram of the structure of the computer device in Embodiment 3 of the present invention. Detailed Implementation

[0058] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of this invention. Obviously, the described embodiments are only some embodiments of this invention, not all embodiments. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention. In the following embodiments, each embodiment provides optional features and examples. The various features described in the embodiments can be combined to form multiple optional solutions. Each numbered embodiment should not be regarded as only one technical solution.

[0059] Example 1

[0060] Figure 1 The flowchart below shows the image registration method provided in Embodiment 1 of the present invention. This embodiment is applicable to the case of local registration of regions of interest in medical images. The method can be implemented by an image registration device, which is configured in a computer device capable of acquiring medical images to be processed. Specifically, it can be implemented through software and / or hardware in the device.

[0061] like Figure 1 As shown, the image registration method specifically includes:

[0062] S110. Obtain the image group to be registered, and perform first-stage registration of the non-reference images in the image group to the reference images in the image group to obtain the preliminary registered image group.

[0063] The image group to be registered includes a reference image and at least one non-reference image. The reference image and non-reference images can be different image sequences obtained at different times and using different scanning sequences for the same scanned object, and the image reconstruction reference coordinates of each image sequence are different. Different image sequences can reflect the characteristics of the scanned object from different aspects or angles. For example, in magnetic resonance imaging (MRI) sequences of head and carotid artery plaques, various bright and dark blood sequences are included, such as 3D TOF (3D Time of Flight Magnetic Resonance Angiography), T1WI (T1 weighted-imaging), T2WI (T2 weighted-imaging), T1CE (T1 contrast-enhanced imaging), and CE-MRA (contrast-enhanced Magnetic Resonance Angiography). These sequences can display the characteristics and properties of plaques from different perspectives, thus facilitating the analysis and evaluation of different plaque types and providing accurate and reliable imaging evidence for clinical practice. Therefore, when analyzing images from different sequences, it is necessary to first perform registration and alignment between the multiple sequences and analyze the images in the same reference space to ensure the accuracy of the analysis results.

[0064] In a group of images to be registered, one image can be randomly selected as the reference image, and the other images as non-reference images. The non-reference images are then registered to the reference image. In a preferred real-time mode, an image with a clear field of view and good image quality can also be selected as the reference image, thereby registering the other non-reference images to the reference image. Specifically, the reference image remains stationary and can also be called the reference image; the non-reference images are transformed and are called floating images. The space formed by using the origin of the reference image as the origin of the coordinate system and the three-dimensional direction vector of the reference image as the coordinate axes of the coordinate system is the space of the reference image. The image registration process involves using a registration algorithm to obtain the spatial transformation relationship (i.e., the registration relationship) between the reference image (reference image) and the floating image (non-reference image), and then resampling and reconstructing the floating image using this spatial transformation relationship to obtain a new floating image (updated non-reference image) that is consistent with the anatomical coordinates of the reference image.

[0065] In the first stage of the registration process in this embodiment, a preset image registration algorithm can be used to determine the registration relationship between the non-reference images and the reference images in the image group to be registered; then, the non-reference images in the image group to be registered are reconstructed according to the registration relationship. The preset image registration algorithm can be a method based on pixel coordinate information, a medical image registration algorithm based on pixel grayscale information, or a medical image registration algorithm based on image features. When using a medical image registration algorithm based on pixel grayscale information, the similarity measure in the algorithm can be mutual information, cross-correlation coefficient, or mean squared error.

[0066] S120. Perform target anatomical structure segmentation on any image in the image group to be registered to obtain the segmentation result of the target anatomical structure in the calibration space where the reference image is located.

[0067] After the first stage of registration, the updated non-reference image and the anatomical structures in the reference image are essentially spatially aligned globally. Image segmentation is performed on any image in the group to be registered, yielding the segmentation results compared to each image in the initially registered image group. These segmentation results can be represented by a mask and / or centerline of the target anatomical structure. In some implementations, the corresponding centerline point set can be obtained by processing the mask. The target anatomical structure can be a tubular structure such as a blood vessel or trachea. Target anatomical structure segmentation can be achieved using region growing methods, level set algorithms, or deep learning-based image segmentation neural networks.

[0068] Specifically, when the target anatomical structure segmentation object is a reference image, the mask and / or centerline of the target anatomical structure obtained from the image segmentation results are the mask and / or centerline of the target anatomical structure in the calibration space where the reference image is located.

[0069] In a preferred embodiment, if the target anatomical structure has higher recognition in a non-reference image, the non-reference image can be used as the object for target anatomical structure image segmentation. A preset image segmentation algorithm can be used to segment the target anatomical structure in the non-reference image, and the mask and / or centerline of the target anatomical structure can be extracted based on the segmentation results. Then, further, according to the registration relationship in the first-stage registration process, the mask and / or centerline are reconstructed to obtain the mask and / or centerline of the target anatomical structure in the calibration space of the reference image.

[0070] S130. The segmentation results are overlaid with each image of the preliminary registered image group and the region of interest in the target anatomical structure is obtained.

[0071] After extracting the mask and centerline of the target anatomical structure, the mask and / or centerline can be overlaid on each image in the initial registered image group. This can be done by using multi-planar reconstruction techniques to display transverse, coronal, and sagittal planes, or by using surface reconstruction techniques to display a stretched image of the target anatomical structure. Multiple images in the initially registered image group can be viewed in a linked manner (simultaneous zooming, panning, and cursor association). Furthermore, the image display area allows users to set, select, and modify parameters such as the start and end points, length, and width of the region of interest (ROI) of the target anatomical structure. This responds to user image selection operations in the image display area, defining the ROI within the target anatomical structure.

[0072] In another implementation, parameters such as the start point, end point, length, and width of the region of interest can also be automatically calculated by a preset target region recognition algorithm and displayed on the software interface. For example, in a magnetic resonance imaging sequence of head and carotid artery plaques, when displaying the reconstructed image of the head and carotid artery after preliminary registration, the blood vessel and coordinate position of the plaque can be identified and located by an automatic plaque detection method, and the blood vessel segment of interest, i.e. the region of interest, can be automatically set according to the plaque position.

[0073] S140. Perform region registration between the region of interest of the non-reference image in the preliminary registered image group and the region of interest of the reference image, and perform second-stage registration on the non-reference image in the preliminary registered image group according to the registration relationship in the region registration process, thereby completing the image registration process.

[0074] Specifically, in the second stage of registration, a preset medical image registration algorithm is first used to determine the region registration relationship between the region of interest (ROI) of each non-reference image in the preliminary registration image group and the ROI of the reference image. Then, based on this region registration relationship, the non-reference images in the preliminary registration image group undergo second-stage registration, completing the image registration process. The preset medical image registration algorithm includes both pixel grayscale information-based and image feature-based algorithms. The registration relationship refers to the rigid transformation matrix, affine transformation matrix, or deformation field determined by the preset medical image registration algorithm.

[0075] After the second stage of registration, the local regions of interest of the target anatomical structure in the image group to be registered are also registered, which improves the accuracy of feature analysis of the region of interest to a certain extent and reduces the error caused by spatial mismatch.

[0076] In a specific magnetic resonance patch analysis, multiple sequences (TOF, T1, T1CE, and T2) with both bright and dark blood were used for joint analysis. The process of spatial registration and alignment between the multiple sequence images is as follows:

[0077] First, for the four sequences TOF, T1, T1CE, and T2, T1 is used as the reference image (baseline image), and TOF, T1CE, and T2 are used as floating images (i.e., non-baseline images). The floating images are then registered to the space containing the reference image. Since the four sequences were acquired during the same inspection process, a rigid registration matrix can be calculated using the coordinate information recorded in each sequence. Using this rigid registration matrix, the floating images are registered to the reference image space. The registration result is as follows: Figure 2 As shown, there is a clear mismatch, indicating that using coordinate information alone often cannot yield correct registration results.

[0078] Furthermore, for the aforementioned multi-sequence data, a mutual information-based image registration algorithm is used for registration to obtain a rigid registration matrix. The floating image updated with the rigid registration matrix determined by the coordinates is then registered to the reference image space using the rigid registration matrix obtained from the mutual information-based image registration algorithm. The registration result is as follows: Figure 3 As can be seen, the macroscopic structures such as the body surface, bones, and muscles in the image have achieved correct spatial alignment with the reference image. However, there is misalignment in some local vascular structures. This is because blood vessels are widely distributed and have tortuous paths. In addition, the microenvironment of blood vessels changes due to the different imaging times of multiple sequences, resulting in complex registration relationships between blood vessels in multiple sequences and failing to satisfy the global rigid transformation relationship. Furthermore, there are factors related to limb movement. For example, in head and neck scans, due to the flexibility of the cervical spine, the relative positions of the head and neck may have changed during the acquisition of multiple sequences. The registration in the above two steps is equivalent to the first-stage registration in step S110. It should be noted that in other embodiments, only one of the above two registration steps can be used as the first-stage registration.

[0079] Furthermore, vascular structures can be segmented using region growing techniques, and specific vascular segments of interest can be selected. For details, please refer to [reference needed]. Figure 4 The vessel segment selection shown is based on the vessel centerline. Figure 4 The implementation involves selecting the start and end points of a vessel segment along the centerline in the surface reconstruction map. The user manually selects the vessel segment to be evaluated (e.g., the segment containing the plaque) as the segment of interest. Next, the second-stage registration can be performed, and the registration result is as follows: Figure 5 and Figure 6 This shows that the local mismatch phenomenon has been significantly improved, reducing the error caused by spatial mismatch in the subsequent assessment of vascular wall parameters and plaque analysis. Figure 5 This is the second-stage registration result for the neck vessel segment of interest. Figure 6 This is the second-stage registration result of the intracranial vessel segment of interest.

[0080] The technical solution of this embodiment involves performing a first-stage registration of non-reference images in the image group to be registered to a reference image, resulting in a preliminary registered image group, achieving spatial alignment at the macroscopic structural level. Then, the target anatomical structure is segmented in any image of the image group to be registered, obtaining a mask and centerline of the target anatomical structure in the calibration space of the reference image. The mask and centerline are then overlaid on each image of the preliminary registered image group to obtain the region of interest (ROI) within the target anatomical structure. Region registration is performed between the ROI of the non-reference images in the preliminary registered image group and the ROI of the reference image. A second-stage registration is then performed on the non-reference images in the preliminary registered image group based on the registration relationship during the region registration process, completing the image registration process. Particularly in the registration process of magnetic resonance images of the head and carotid arteries, this method can effectively improve the registration effect of local blood vessels, thereby increasing the accuracy of plaque analysis within the segment of interest. This embodiment of the invention solves the problem of poor registration effect in local regions that do not meet rigidity requirements; it can improve the registration effect of local image regions, thereby increasing the accuracy of ROI analysis.

[0081] Example 2

[0082] Figure 7 This is a schematic diagram of the structure of an image registration device provided in Embodiment 4 of the present invention, which can be applied to various situations.

[0083] like Figure 7 As shown, the image registration device in this embodiment of the invention includes: a first registration module 210, a structure segmentation module 220, a target region selection module 230, and a second registration module 240.

[0084] The system comprises the following modules: a first registration module 210, which acquires a group of images to be registered and performs a first-stage registration of non-reference images in the group of images to be registered to a reference image in the group of images to be registered, thereby obtaining a preliminary registered image group; a structure segmentation module 220, which performs target anatomical structure segmentation on any image in the group of images to be registered, thereby obtaining the segmentation result of the target anatomical structure in the calibration space where the reference image is located; a target region selection module 230, which overlays the segmentation result with each image in the preliminary registered image group and acquires the region of interest in the target anatomical structure; and a second registration module 240, which performs region registration of the regions of interest of the non-reference images in the preliminary registered image group with the regions of interest of the reference image, and performs a second-stage registration of the non-reference images in the preliminary registered image group according to the registration relationship in the region registration process, thereby completing the image registration process.

[0085] The technical solution of this embodiment involves performing a first-stage registration of non-reference images in the image group to be registered to a reference image, resulting in a preliminary registered image group, achieving spatial alignment at the macroscopic structural level. Then, target anatomical structure segmentation is performed on any image in the image group to be registered, yielding a segmentation result of the target anatomical structure in the calibration space of the reference image. This segmentation result is then overlaid with each image in the preliminary registered image group to obtain the region of interest (ROI) within the target anatomical structure. Region registration is performed between the ROI of the non-reference images in the preliminary registered image group and the ROI of the reference image. A second-stage registration is then performed on the non-reference images in the preliminary registered image group based on the registration relationship during the region registration process, completing the image registration process. This embodiment of the invention solves the problem of poor registration results in local regions that do not meet rigid change requirements; it can improve the registration effect of local image regions, thereby increasing the accuracy of ROI analysis.

[0086] Optionally, the first registration module 210 is specifically used for:

[0087] The registration relationship between the non-reference image and the reference image in the image group to be registered is determined according to a preset image registration algorithm;

[0088] The non-reference images in the image group to be registered are reconstructed according to the registration relationship.

[0089] Optionally, when the object of image segmentation is a non-reference image in the group of images to be registered, the structure segmentation module 220 is specifically used for:

[0090] A preset image segmentation algorithm is used to segment the target anatomical structure of the non-reference image in the image group to be registered, and the segmentation result of the target anatomical structure is extracted.

[0091] Based on the registration relationship in the first stage registration process, the segmentation result is reconstructed to obtain the segmentation result of the target anatomical structure in the calibration space where the reference image is located.

[0092] Optionally, the target area selection module 230 is used for:

[0093] A preset target region recognition algorithm is used to detect the region of interest in the target anatomical structure.

[0094] Optionally, the target area selection module 230 is further configured to:

[0095] In response to an image selection operation in the image display area, a region of interest is determined in the target anatomical structure.

[0096] Optionally, the preset image registration algorithm includes:

[0097] Methods for calculating based on pixel coordinate information, and medical image registration algorithms based on pixel grayscale information or image features, wherein the similarity measure in the medical image registration algorithm based on pixel grayscale information includes mutual information, cross-correlation coefficient, or mean square error.

[0098] Optionally, the second registration module 240 is specifically used for:

[0099] By using a preset medical image registration algorithm, the region registration relationship between the region of interest of the non-reference image in the preliminary registration image group and the region of interest of the reference image is determined. Then, based on the region registration relationship, the non-reference image in the preliminary registration image group is registered in the second stage to complete the image registration process.

[0100] The image registration apparatus provided in this embodiment of the invention can execute the image registration method provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the method.

[0101] Example 3

[0102] Figure 8 This is a schematic diagram of the computer device in Embodiment 3 of the present invention. The computer device can be connected to an imaging device (such as a CT, PET, or MRI device) to control the imaging device, receive signals collected by the imaging device, and process the collected signals. Alternatively, it can obtain image data to be processed through a network or by accessing a storage device. Figure 8 A block diagram of an exemplary computer device 12 suitable for implementing embodiments of the present invention is shown. Figure 8 The computer device 12 shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of the present invention.

[0103] like Figure 8 As shown, the computer device 12 is represented in the form of a general-purpose computing device. The components of the computer device 12 may include, but are not limited to: one or more processors or processing units 14, system memory 28, and a bus 18 connecting different system components (including system memory 28 and processing unit 14).

[0104] Bus 18 represents one or more of several bus architectures, including a memory bus or memory controller, a peripheral bus, a graphics acceleration port, a processor, or a local bus using any of the various bus architectures. For example, these architectures include, but are not limited to, the Industry Standard Architecture (ISA) bus, the Micro Channel Architecture (MAC) bus, the Enhanced ISA bus, the Video Electronics Standards Association (VESA) local bus, and the Peripheral Component Interconnect (PCI) bus.

[0105] Computer device 12 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by computer device 12, including volatile and non-volatile media, removable and non-removable media.

[0106] System memory 28 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 30 and / or cache memory 32. Computer device 12 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, storage system 34 may be used to read and write non-removable, non-volatile magnetic media (…). Figure 8 Not shown; usually referred to as a "hard drive"). Although Figure 8 Not shown, a disk drive for reading and writing to a removable non-volatile disk (e.g., a "floppy disk") and an optical disk drive for reading and writing to a removable non-volatile optical disk (e.g., a CD-ROM, DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to bus 18 via one or more data media interfaces. Memory 28 may include at least one program product having a set (e.g., at least one) of program modules configured to perform the functions of the embodiments of the present invention.

[0107] A program / utility 40 having a set (at least one) of program modules 42 may be stored, for example, in memory 28. Such program modules 42 include, but are not limited to, an operating system, one or more application programs, other program modules, and program data. Each or some combination of these examples may include an implementation of a network environment. Program modules 42 typically perform the functions and / or methods described in the embodiments of the present invention.

[0108] Computer device 12 can also communicate with one or more external devices 14 (e.g., keyboard, pointing device, display 24, etc.), and with one or more devices that enable a user to interact with the computer device 12, and / or with any device that enables the computer device 12 to communicate with one or more other computing devices (e.g., network card, modem, etc.). This communication can be performed via input / output (I / O) interface 22. Furthermore, computer device 12 can also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN), and / or public networks, such as the Internet) via network adapter 20. As shown, network adapter 20 communicates with other modules of computer device 12 via bus 18. It should be understood that, although... Figure 8As not shown, it can be used in conjunction with computer device 12 with other hardware and / or software modules, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0109] Processing unit 16 executes various functional applications and data processing by running programs stored in system memory 28, such as implementing the image registration method provided in this embodiment of the invention, which mainly includes:

[0110] A group of images to be registered is obtained, and the non-reference images in the group of images to be registered are registered to the reference images in the group of images to be registered in the first stage to obtain a preliminary registered image group.

[0111] Perform target anatomical structure segmentation on any image in the group of images to be registered to obtain the segmentation result of the target anatomical structure in the calibration space where the reference image is located;

[0112] The mask and / or centerline are overlaid on each image of the preliminary registered image group to obtain the region of interest in the target anatomical structure.

[0113] The regions of interest (ROIs) of the non-reference images in the preliminary registered image group are registered with the ROI of the reference image. Then, based on the registration relationship in the ROI process, the non-reference images in the preliminary registered image group are registered in a second stage to complete the image registration process.

[0114] Example 4

[0115] Embodiment 4 of the present invention also provides a computer-readable storage medium storing a computer program thereon. When executed by a processor, the program implements the image registration method provided in the embodiments of the present invention, which mainly includes:

[0116] A group of images to be registered is obtained, and the non-reference images in the group of images to be registered are registered to the reference images in the group of images to be registered in the first stage to obtain a preliminary registered image group.

[0117] Perform target anatomical structure segmentation on any image in the group of images to be registered to obtain the segmentation result of the target anatomical structure in the calibration space where the reference image is located;

[0118] The segmentation results are overlaid with each image in the preliminary registered image group to obtain the region of interest in the target anatomical structure.

[0119] The regions of interest (ROIs) of the non-reference images in the preliminary registered image group are registered with the ROI of the reference image. Then, based on the registration relationship in the ROI process, the non-reference images in the preliminary registered image group are registered in a second stage to complete the image registration process.

[0120] The computer storage medium of this invention can be any combination of one or more computer-readable media. A computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. For example, a computer-readable storage medium can be, but is not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this document, a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0121] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, capable of sending, propagating, or transmitting programs for use by or in connection with an instruction execution system, apparatus, or device.

[0122] Program code contained on a computer-readable medium may be transmitted using any suitable medium, including but not limited to wireless, wire, optical fiber, RF, etc., or any suitable combination thereof.

[0123] Computer program code for performing the operations of this invention can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0124] Note that the above description is merely a preferred embodiment of the present invention and the technical principles employed. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described herein, and various obvious changes, readjustments, and substitutions can be made without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in detail through the above embodiments, the present invention is not limited to the above embodiments, and may include many other equivalent embodiments without departing from the concept of the present invention, the scope of which is determined by the scope of the appended claims.

Claims

1. An image registration method, characterized in that, include: A group of images to be registered is obtained, and the non-reference images in the group of images to be registered are registered to the reference images in the group of images to be registered in the first stage to obtain a preliminary registered image group; wherein, the group of images to be registered is a different image sequence obtained at different times and using different scanning sequences for the same scanned object; Perform target anatomical structure segmentation on any image in the group of images to be registered to obtain the segmentation result of the target anatomical structure in the calibration space where the reference image is located; wherein, the segmentation result is represented by a mask and / or center line of the target anatomical structure; The segmentation results are overlaid with each image in the preliminary registered image group to determine the region of interest in the target anatomical structure. By using a preset medical image registration algorithm, the region registration relationship between the region of interest of the non-reference image in the preliminary registration image group and the region of interest of the reference image is determined. Then, based on the region registration relationship, the non-reference image in the preliminary registration image group is registered in the second stage to complete the image registration process.

2. The method according to claim 1, characterized in that, The first-stage registration of the non-reference images in the image group to be registered to the reference images in the image group includes: The spatial transformation relationship between the non-reference image and the reference image in the image group to be registered is determined according to a preset image registration algorithm and used as the registration relationship. Based on the registration relationship, the non-reference images in the image group to be registered are reconstructed to obtain updated non-reference images, such that the updated non-reference images are in the same reference space as the reference images.

3. The method according to claim 1, characterized in that, When the object of image segmentation is a non-reference image in the image group to be registered, the step of segmenting the target anatomical structure of any image in the image group to be registered, and obtaining the segmentation result of the target anatomical structure in the calibration space where the reference image is located, includes: A preset image segmentation algorithm is used to segment the target anatomical structure of the non-reference image in the image group to be registered, and the segmentation result of the target anatomical structure is extracted. Based on the registration relationship in the first stage registration process, the segmentation result is reconstructed to obtain the segmentation result of the target anatomical structure in the calibration space where the reference image is located.

4. The method according to claim 1, characterized in that, The step of overlaying the segmentation results with each image of the initially registered image group includes: Based on the mask and / or centerline of the target anatomical structure, the mask and / or centerline of the target anatomical structure are overlaid on each image of the preliminary registered image group.

5. The method according to claim 1, characterized in that, The step of acquiring the region of interest in the target anatomical structure includes: A preset target region recognition algorithm is used to detect the region of interest in the target anatomical structure.

6. The method according to claim 1, characterized in that, The step of obtaining the region of interest in the target anatomical structure further includes: In response to an image selection operation in the image display area, a region of interest (ROI) is determined in the target anatomical structure; wherein the image selection operation includes editing the start and end points of the ROI in the target anatomical structure.

7. The method according to claim 2, characterized in that, The preset image registration algorithm includes: Methods based on pixel coordinate information calculation, and medical image registration algorithms based on pixel grayscale information or image features.

8. An image registration device, characterized in that, include: The first registration module is used to acquire a group of images to be registered, and to perform a first-stage registration of the non-reference images in the group of images to be registered to the reference images in the group of images to be registered, thereby obtaining a preliminary registered image group; wherein, the group of images to be registered is a different image sequence obtained at different times and using different scanning sequences for the same scanned object; The structure segmentation module is used to segment the target anatomical structure of any image in the group of images to be registered, and obtain the segmentation result of the target anatomical structure in the calibration space where the reference image is located; wherein, the segmentation result is represented by a mask and / or center line of the target anatomical structure; The target region selection module is used to overlay the segmentation results with each image in the preliminary registration image group to determine the region of interest in the target anatomical structure. The second registration module is used to perform region registration between the region of interest of the non-reference image in the preliminary registration image group and the region of interest of the reference image, and to perform a second-stage registration of the non-reference image in the preliminary registration image group according to the registration relationship in the region registration process, thereby completing the image registration process. Specifically, the second registration module is used for: By using a preset medical image registration algorithm, the region registration relationship between the region of interest of the non-reference image in the preliminary registration image group and the region of interest of the reference image is determined. Then, based on the region registration relationship, the non-reference image in the preliminary registration image group is registered in the second stage to complete the image registration process.

9. A computer device, characterized in that, The computer device includes: One or more processors; Storage device for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the image registration method as described in any one of claims 1-7.

Images